Research On Calibration And Accuracy Of Laser Scanner In Articulated Coordinate Measure Machine

Reverse engineering (RE) is one main branch of advanced manufacture technology currently and plays an important role by quickly absorbing advanced manufacturing technology, shortening development cycle and cost reduction. Meanwhile, Articulated Coordinate Measure Machine (ACMM) is becoming a focus in the field of RE because it has lots of advantages, such as small volume, low weight, portable carriage and unlimited measure range by worktable. Nowadays, overseas ACMM technology has been mature, but the machine price is every expensive. Therefore, it's a significant and urgent job to develop a relative mature tool with low cost in China. On the foundation of previous research in this field, following aspects are further studied: calibration and precision of the laser scanner of newly developed ACMM.Firstly, based on the character of ACMM, which is build on the single CCD visual survey technology, the introduction is given to the single CCD vision measuring method. And several common single CCD visual measuring methods are listed out in detail. Then, the selection rules of the measurement system key parts such as lens, CCD and image gathering module card are analyzed.Secondly, the principle of ACMM work and the laser scanner imaging are briefly introduced. For the laser measurements coordinate measuring machine, the single CCD and the structure light are used to match the three-dimensional data of objects'surface. Based on this measurement method, a new ladder-shaped calibration block is designed and two images are taken: one is structured stripe image and the other is laser plane image by this block. The laser plane parameter and internal and external camera parameter calibration can be accomplished by both images.Finally, several influence factors of three dimensional laser scanning system precision are analyzed. According to distortion-free, only radial distortion camera models, camera parameter calibration experiments and the calibration error analysis are performed with 12 and 30 calibration holes separately. And a new calibration method is proposed according to radial distortions, decentering distortion, thin prism distortion. Then camera parameter calibration experiments and the calibration error analysis are performed by the new method based on above radial distortions, decentering distortion, thin prism distortion camera models. Finally, the perspective projection model of circle center in 3D visual inspection and the distortion error model are analyzed, and the corresponding correction method is proposed.